Pro mastering engineer reveals embarrassing mistakes
ฝัง
- เผยแพร่เมื่อ 5 ก.ค. 2024
- ⭐ STEM MASTERING MASTERCLASS - www.panoramamastering.com.au/...
📧 The BEST newsletter for working audio professionals who strive for more! Subscribe here - www.panoramamastering.com.au/...
🎹 Proud of your mix? AWESOME! Get a fresh set of ears on board for mastering and let’s work together! - panoramamastering.com.au
🛠️ Mastering Template for ProTools, Logic and Ableton- www.panoramamastering.com.au/...
--
Hello, I'm Nicholas Di Lorenzo, Studio Owner, Mixing and Mastering engineer at Panorama Studios.
I'm an Italian-Australian born and raised in Melbourne. I've been a creative professional for 10 years managing some pretty awesome projects for artists, labels and producers all around the globe.
What motivates and drives me?
My family,
Good food,
Great coffee.
You can find me on many platforms:
Instagram: / panorama_mastering
Facebook: / panoramamastering
Twitter: / panoramamasters
Kit: kit.co/Panorama_Mastering
Better dither explanation i've EVER heard, Thanks man u are the best
You're kinda saying the good ol' "use ur mf ears" but in the most tasteful way. Thanks for putting so much effort into ur vids and currently being one of the most honest YT audiopros. cheers
Glad you like them! I am enjoying making these
"I'm not sure if that hand action works..."
I was dying.
I was going to edit visual drawings for those explanations but the hands did it all
@@panorama_mastering Nah brother, strange gestures and unintentional humor are definitely the way forward. It was unexpected and caught me in just the right moment to send me into a fit of hysterical laughter. Aside from all the great content you put out, thank you for making me laugh. Still one of the best sounds ever. Not me, mind. Just laughing in general.
Thx, nice video. Regarding the last point, there's also another thing to take into account. Just "clipping plugins" is kinda of a strange statement. Clippers have their own architecture, but too many plugins don't react well to clipping, specially on the output. Stillwell Audio still has that Bitter tool for measuring Bit Depth of plugins.
A breath of fresh air in the sea of lazy audio content that's plaguing youtube nowadays. Finally I don't feel like the video is fighting to keep my attention and just genereate meaningless watchtime. Genuinely interesting video, learned some stuff, great pace, good demonstrations, love it!
Great video awesome content thank you for always educating and being transparent with your knowledge and experience
My pleasure! Glad to have you aboard :)
Cheers Nicholas!
Best channel on YT
Thanks man! My favourite at the moment has to be Dan Worralls!
My understanding is that dither is not simply added noise. It results in “randomisation of the least significant bit” of the truncated signal.
🤓
The noise is added before truncation. It allows to recover a signal that would have been reduced to complete silence because it causes the noise samples to move up and down following the signal.
The compression releace and attack times get even more weird, interesting, and "unpredictable" (unless you're like a rocket surgeon or something lol) when you start digging into how optical compressor work on an electrical level! (especially the older non LED ones).
Spot pn, I did a video discussing the topologies. th-cam.com/video/sQt-vMEJ0gI/w-d-xo.htmlsi=S4t4THI-J-E4Z2Q4
Have you touched on feed back and feed forward compressors in any vid?
It's something not a lot of people are aware is even a thing, and can have a huge impact on the way a compressor reacts, and can change the sound of it considerably even between two of the same types of compressor (even the exact same compressor with both feedback and feedforward modes will react and sound diferant in each mode) .
Even if just a small section of a vid, it'd be something I'd be interested in. I know how it works, but you always manage to find some nugget of info I've not heard before, so am interested in hearing your explanation of it.
I know it might not be the most important thing to worry about in mastering, but it's still good to know about I think.
So oversample before compression? It makes perfect sense like limiting a vocal for a stronger signal before certain processing. Thank you I have to try this
DMG Compassion. If you haven't dig it already, I think I learn attack release relation a lot from that plugin. I also love how it has noise function to add to the side-chain circuit to blur out the detection to mimic the analog style compressor imperfection. It on the expensive side of things however.
Some good nuggets in there. I had never seen that arithmetic for attack and release before. I had the same prior assumptions as you.
I've noticed the low end weirdnesses, now I know why. Thanks Nick. What's that Worrall video link?
th-cam.com/video/1ormfTMYfv0/w-d-xo.htmlsi=F4r1zdd4dXT13xRs
So, you always start mastering with a Clipper, if you have a track where you for some reason need to filter the low end. Would you use a linear phase Eq after the clipper still? I noticed you had the Pro Q3 on zero latency. Or would you just filter with a Eq, the clipper, then more EQ. Your words were aggressive Eq. So I wonder if with the normal 0.5 to 2.0 dB mastering movements in a Eq it will also affect the wave form that bad. What would you do in that situation?
Clipper then EQ(Linear phase) or EQlinear phase then Clipper. Thank you so much, sorry if I misunderstood something.
Hey thanks for a great video. This should have been 5 different videos on the subjects thoroughly explained but we appreciate. It means I will have watch the video multiple times to understand. Thank you very much.
Haha good idea!
time constant of 100 ms means after 100 ms the attenuation is completed by 63 percent. as our loudness perception is logarithmic as well id think though that this effect is antiproportional to the amount of compression. the less gain reduction the higher this impact will be. so there is no big difference in loudness between one time constant and 5 time constants(5*100ms)which is considered 100 percent attenuation when compressing by 10 dB. and it will be more audible when compressing by 2-3 dB.
Yes, that first time constant the gain reduction is at it’s greatest velocity
Great stuff! Do you know if this time constant is inherent to all compressors in equal amounts, or does it vary, especially ITB?
It would depend on the type of compressor in some cases, This is why some compressors are naturally fast types and others naturally slow. As they all acheive the compression effect in diferant ways, what is causing the attack and releace times to be what they are can vary. For example a super fast VCA compressor is all IC chip based and would rely on capasitor discharge time quite heavily, where as an optical compressor is a light shining on a light sensitive variable resistor, and is inherently slow to respond as a result (they also don't have a liniar or logarythmic responce afaik, at least not for their whole curve), and not only vary their releace curves over time, sometimes quite unpredictibley, they also have a memory! As a result they are program dependant too and will react diferantly to diferant things you send through them... O.o
With regards to ITB, most standard compressors that arn't emulating gear would *most likely* be exactly as Nick describes, however there's a bunch of emulations that mimik hardware aswell, including all the diferant way of compressing in analouge. However things like most DAWs stock compressor for sure would most likely be strictly logarythmic as Nick described (and some of the emulations probably are aswell).
Back in the analoug realm, they would all somewhat be affected by the capasitor discharge time, but only predominantly if the capasitor discharge time is the "weakest link" and the slowest componant to react to and change within the gain reduction circuit (like they are in most modern super fast compressors). They still have an effect on things even if they are not the slowest thing to react, however it's significance in determining the A and R curve is greatly reduced, as the slowest componant to react in the circuit will impart it's discharge or equivilent charicteristics into the releace time instead (again like with optical compressors where the LDR will get "exiceted" by high or long levels of light way after the light has gone off, and any subsiquent short burst of light will then be effected by the memory of the cell of that bright or long exposure to light which will be long after any of the caps within that circuit take to discharge in most cases I would think).
Depends on the type of compressor and circuit.
Cool!
Nice aproaching.
Talking about clippers...
What about those ones that use clippers just to shave transient spikes (mostly from percussive instruments)?
I prefer clippers to get my tracks as loud as possible shaving the transient spkies before the stereo master and another one in the stereo master set to as hard as possible at 0dB steady as a rock.
The problem is the CPU if you try to use oversampling in too many tracks.
So, I know the subject is not the mixing part of the job, but I think that, maybe, it could help a bit to get a better mastering stage...
Yes! Clippers are a great tool, (hard clippers for short transients)
i need this tee so bad, where did you get it
Exclusive Panorama drip!
favourite dither? goodhertz or MAAT?
Wait so you always dither when exporting your original lossless file? Or when converting from that OG file to a lower bit rate?
Always. Even og lossless master, never know what people will go and do with that file, rather have it on than not
@@panorama_mastering Allright thanks for answering!
9:59 so this is why we get the *knee on compressors? Ty
So does it mean that clip drum bus is not good, because you will have eq on 2bus? A lot of engineers clip kick, then drum bus then 2bus, but it is 100% that will be some eq action after clipping
Depends on the EQ action. Big boosts/cuts with shelves or using pass filters will shift the peak noticably.
Minor EQ tweaks and broad bells won’t do so noticeably
Another awesome video from my man, Nick! I think we need a follow-up video on the K-filtering thing. My understanding is that the purpose of K-filtering is to provide a more accurate representation of how humans perceive loudness. So why is everything below 100 Hz being filtered? I know low-end has an imbalance when it comes to energy vs perceived loudness, but is it necessary to eliminate it altogether like that? Not to mention there’s an additional high shelf boost on top of the high pass filter. (Side note: why did you call the shelf a second-order high pass? Is that because it has a 12 dB slope?). Anyway, just curious how all that K-weighting stuff works. I’m not questioning the validity of the filter but rather why such a drastic removal of low end is necessary to accurately measure loudness. I would think that the filtering would be more representative of an inverted Fletcher Munson curve. 🤷 😂.
(I typed this out and thought it might come across as rude, I didn't mean it like that I swear! 🤓)
12 dB/oct is a second order filter, 6dB per order, so to speak. 12dB/oct at 100hz doesn't mean everything is eliminated altogether, it means it gets lowered 12dB per octave (down 3dB at 100hz). Of course, the K-filter isn't exactly analogous to any (inverted) equal loudness contour, as that would make the filter very complex, computationally (or electronically). K-filter is easy in that regard, and also "close enough", with its high pass filter and top end boost. And equal loudness contours are just squiggles on graph paper anyway, an average of measurements of a random sample of people. There is very little to be gained by exactly tracing an equal loudness contour with a filter, especially as the use of this filter in this use case only amounts to helping a computer to spit out a loudness measurement. And not to be one of those "the song is all that matters, man" types, but the different number doesn't make it a better or worse song.
That explanation wasn't rude at all, haha! It pretty much clears things up for me. So, basically it’s a straightforward and "good enough" way to filter the audio for accurate measurements, right? I had a feeling that might be the case. I think the visuals of the filters were throwing me off, but now that you've explained the function of a second-order filter, it makes more sense. So, with this logic, at 50 Hz, you're around -12 dB lower from the original volume, and at 25 Hz, it's about -24 dB, and so on. Makes a lot more sense now. Thank you!
I will do!
Yes second order is 12dB/oct
Their goal was to make a super simple filter, so they only used two "parts". Various more complex loudness measurement systems already existed at the time that were "not invented here". The loudness measurement runs much faster than, say, the one you get in Sound Forge where you have an equal loudness contour. It was also made primarily for TV where they don't care about response at 20 Hz.
Can you give a short summary of lesson number 2 (low cut and clippers)? If i understand correctly if we need to high pass we should do it before clippers or bette,r don''t do any high passing as it increases peak values?
If you need to do it, do it. It's better to not do it into anything that shifts phase such as an EQ if needing to do heavy equing or hp/lp etc as it will rotate the phase of your clipped signal and shift the flat tops to be on the side of the wave instead, seens as the phase has been rotated by the EQ etc. Using a HP/LP may or may not increase peaks, as far as I understand it it would depend what way the signal is rotated in phase at the peak, as one way would cause more gain, and the other less. I'm not sure if there's any way to work that out easily when EQing, so it's probably best to be mindful and caucious when doing any heavy EQ/filtering and just do it before the clipper, that way even if it adds gain, the clipper can deal with it, and because it's before the clipper, the clipping doesn't get it's phase rotated too and you retain your flat tops.
Or use Linnear Phase eq.
@@rossbalch Linnear eq introduce pre ring and so on and so forth. I am sure that he was trying to make a point but I don't get it
Correct hpf before clippers
Nerdy Nerdy 😢🤯🤯. I CAN’T BREATHE !
First time ive landed here... Good Stuff... New Sub.
oi oi oi
Welcome aboard! M808!!!
I usually use proQ3 in linear phase when in mastering - if doing anything more then like a db.... Its funny cause Ive never worry/ think about phase when using my analog gear lol I also wonder if your client actually heard something or just saw this and made that comment ( nothing wrong with that) I know ive heard tracks that i thought sounded pretty damn good but had overs ( mostly dance music though)
I am curious if the prering will shift the clipping at all? Time to test!
@@panorama_mastering if you make a video ill be watching :) i also almost always do a little clipping first so id be interested to see the results !!
4:48 I think you forgot to add the link to dan worrals video
Here you are!!
th-cam.com/video/1ormfTMYfv0/w-d-xo.htmlsi=7VDjfTcK7lsYWXVp
Just a few mistakes about the range of integers for a particular bit depth, and about the noise floor level.
For a bit depth 'n', the range of integers being represented is [-2^(n-1)] to [2^(n-1)-1].
For 8 bit, that's -128 to 127.
For 16 bits, that is -32768 to 32767.
Etc.
Also, since we’re dealing with signed integers, the noise floor lies at around -42dBFS at 8 bit bit-depth, -90dBFS at 16 bits, etc.
Good observations! Thanks for clarifying!
Need that shirt merch though fam.
How's this instead; I do a limited drop and give them away for free to 5 people?
Would you be interested?
@@panorama_masteringI’d be interested that shirt is great
@@panorama_mastering say less fam.
Please do a livestream talk with Paul Ashmore from Audio Animals. He claims that analog mastering is way superior in sound quality than ITB mastering only and I'd love to see you both talk about that haha
Maybe!
@@panorama_mastering that would make a cool video!
My brain hurts....
Mine too!
Well now IM EMBARRASSED for struggling to understand a word you just said. Lol. Im a producer that can mix in certain genres but want to understand the mastering process more and how to read meters better when scrolling through, peak, rms, lufs. What's the best place to start learning this foreign language ?
Dan Worral
Shouldn‘t you just dither when lowering the bit rate when exporting? So if the DAW project is 24-bit and you export to 24-bit wav, mp3, m4a, … no dither is required. Or do you add always dither in case the streaming services lower the bit rate?
Nice Video btw! 🙂
Oh and I mean bit depth, not bit rate of course 😅
Most DAWs use 32 or 64 bit floating point.
Dither should theoretically be used due 24bit export.
I would be surprised if Anyone could here a difference though.
Thanks, yes you‘re right, so theoretically dither should be applied 🙂
damn...
I had no idea what you were talking about at any point in this video. Most of your videos I understand, but this was Star Trek techno babble.
I'm 90% sure you're saying "aliasing" wrong for engagement 🤣. There is absolutely no reading of that word the results in "Ay-lee-eye-sing"
(currently my favourite yt channel for mastering btw... keep ay-lee-eye-sing your way to the top)
Haha cheers! Honestly, I have no clue how to say it… even when people catch me on calls correcting me… showing me… i canmt hack it
@@panorama_masteringAye Lee ass (or ess) ing, depending on which state in Australia you grew up in.
Today experience for me
If i 1db amount of clip on master track
It will reduce 1 amount of dynamic range....
Amazing! Thanks for sharing
I
Yes!!
Don't be so hard on yourself mate :)
☻ This is just amazing to know!!!!
You’re welcome!
To this day I still don't understand why there are master engineers? What's the point? Why doesn't the mixing engineer do it right away? I do not get it.
The purpose of a mastering engineer is to put the final polish on a mix, ensuring that it sounds consistent, balanced, and optimized for various playback systems. While mix engineers focus on blending individual tracks within a song, mastering engineers work on the entire album or project, ensuring continuity across tracks, adjusting levels, applying equalization, compression, and other processing to achieve a cohesive sound. Additionally, mastering engineers prepare the final tracks for distribution, making sure they meet industry standards for formats like CD, vinyl, streaming, etc.
Mix engineers could technically attempt mastering, but mastering is a specialized skill that requires a fresh perspective and a different set of tools and techniques. Plus, having a dedicated mastering engineer provides an objective ear and ensures that the final product meets professional standards.
The best and simplest answer is an extra set of professional ears with limited access to make changes.
Producers/artists limit mixers with the choices made within the production
The mix engineer then limits mastering engineer by sending a stereo file (generally)
And the mastering engineer adds the final touches. All the while a new set of ears attached to a new brain all working to make the track as good as possible.
It's the final check in the other room, other system by a very trained ears and final few tweaks to make it sound 5-10% better at the end. If the mix is amazing the mastering engineer confirms that it doesn't need anything and he gets paid just for that. He's the quality judge
Good question! Addressing in a reaction vid soon!
@@j-station Is definitely the best & accurate answer
Please , for French guy who speak english heu .... like a french guy .... Please can you slow down the speed of your sentences ?
Speak more slowly for those far away .... In fact google translator don't like it . hahahaha But your comments are very enriching and enlightening. Thank you .
No sweat! I will try!
Just a reminder........in the past there were excellent mixes and masters without all this ....knowledge and terminology........and don't say its because it was analog.......No its because people were relying on their ears and their experience.My advice ..the most shit like this you follow the worse your sound will be.....friendly :)
Write
a
BOOK
Do it Nick.
Do eeeeeeeeeeeetttttt
I’d love to. But I am affraid my effort can spent in other endeavours helping the community in other ways more effectively
@@panorama_mastering Then I am going to have to just keep watching but not fully understanding 😂
Maths class
1+1=window
another zinger
Zing!
The converter part of your video I would deem embarrassing but the rest is the journey. Great video for new comers to be humble and open. 🫡
Glad you enjoyed it!
First
Banggg
You are doing it wrong, I don't. like these titles at all 🤬
That’s why you clicked on it and watched it. Smart move from Nicholas
What should I have titled this?
@@brucerobertwayne who said I watched the video 🤣
@@panorama_mastering I dunno 🤷🏼♂️
We don’t know other people’s workflow so saying that they are doing it wrong is a assumption. Could be something like
“ maybe you’re doing it wrong “
Actually hè say’s “my former self was wrong and I’m completely fine with exposing my own mistakes so other people can benefit from it” but maybe you should watch the video that helps.
Is it just me or are you mad about numbers? Get a life and use your EARS.
Music is about capturing an emotion. If it sounds good and feels good then it is good. It's easy to fk things up when you overthink art.
Third